Q: Warranty At 800Mhz?


Status
Not open for further replies.
I think we should just pretend this thread didn't happen. The whole overclocking thing has always been nonsense. You can't advertise(actively or not actively is irrelevant, imo) overclocking and say _but_ you void the warranty, this late. Either you accept that someone cheats _if_ something breaks due to overclocking and repair the device under warranty or you block it in the OS. You can't have it all.

I request this thread to be unstickied and let it fall off the first page when everything is said.

Peace! :)
 
Topic unpinned.

@mali, I originally pinned the topic earlier, Because i thought it may answer questions that many would be asking.

I did think it could turn into a bit of a slinging match.

Oh well, I guess we`ll just let the topic fade off into obscurity.

Trooper
 
In the Catalyst Control Center for ATI video cards there is the 'overdrive' ability to overclock your card but it comes locked, and unlocking it (just pressing the button to do so) gives you a warning about your warranty and potential damage. Something like that for the overclocking GUI on Pandora makes concerns of the 'tools provided' nature irrelevant - you could still overclock without the CCC and on Pandora aswell without the GUI but those are not really provided in the same way.

But it's still BS, 600mhz isn't going to be enough for a lot of high-end stuff, in many many threads by developers they say 'here is what's playable at 600' and 'here is what's playable at 800' or some such. There have been many discussions and threads about this but none this harsh. Or, of course, close to Pandora delivery. Which is BS.
 
Hi,
Dead1nside said:
My point rather, was that some companies for instance invalidate your warranty if you run Linux on your laptop, or at least have tried to in the past (I know this is a different issue).
Really ? :blink:
Bye, Magic Sam
 
Last edited by a moderator:
craigix said:
mali said:
Dead1nside said:
Okay so if we do any overclocking in software, without changing the voltages this will no matter what, even if it isn't the cause of any problems, void our warranty?
Nobody will be able to tell if you overclocked it or not, thus it's a non-issue ;)

_If_ it fails, you can either tell them you overclocked or you don't. It depends on you altruism coefficient. I got a replacement motherboard from Asus for my A600, because I didn't tell them.

So you're a dishonest little shit too?

I want your order number. You are banned from having a Pandora from us.

That's right, your order is canceled.
Seems Craig has been watching Minority Report far too many times? :D
 
Last edited by a moderator:
I think Craig should be careful about what he posts on a forum as everything could be considered official information, legally binding etc. since he's selling Pandoras and involved in the development. While perhaps not stated in clear text the Pandora have been put forward as an overclock friendly device by many team members and would be legally binding at least by Swedish consumer laws, this is because the consumers have been made to understand that overclocking is risk free and allowed by warranty before the purchase. By Swedish law it is also not possible for the seller to make any changes to what he is selling after making the offer to sell, nor take his offer back. I do not know about UK laws but I would be surprised if the consumer rights are not similar to ours.

All I'm saying is that Craig and OP should be careful about what they are posting, no matter where. Consumer rights against companies are usually very strong. Our economies are based on consumption and strong consumer rights are there to encourage and protect continuous consumption.
 
Awakening said:
Our economies are based on consummation and strong consumer rights are there to encourage and protect continuous consummation.

Which rights body should I talk to about getting my fair share of this continuous consummation?
I feel I should do my bit for the economy, you understand ;)

http://en.wikipedia.org/wiki/Consummation
 
Last edited by a moderator:
Squidge said:
peca said:
If overclocking functionality is part of preinstalled OS then user can't be sued for destroing HW due to overclocking.
So, using that analogy, the person in this video should be able to claim for a new engine through the car warranty, as the car allowed him to do that damage using the standard controls?
OK. I understand. But:

1/ There was discussion about PND/PXML functionality that allow automatic overclocking for demanded emus, games, etc. I hope that I remember this correctly and not just dreamed about.
2/ I watching this project from xmas 2007 and if I recall it correctly, there was always rumours about overclockability. Some devs say that run emulator on 800MHz flawlessly. I still have something about 900MHz on mind, but really don't know from where it came.

May be this was declined and I just miss something. Please don't take me as some dumb troll. I just trying to find what is still secure and what is risky.

Official pages stays 600Mhz+. So it evoke that overclocking is OK. All those messages about devs Pandoras running on more than 600MHz asure as that it is not something bad.

Seems to some proclamation and user affirmation (warning and agreement) will be needed to use overclocking utility/ability :)
 
Last edited by a moderator:
Asmo said:
Awakening said:
Our economies are based on consummation and strong consumer rights are there to encourage and protect continuous consummation.

Which rights body should I talk to about getting my fair share of this continuous consummation?
I feel I should do my bit for the economy, you understand ;)

http://en.wikipedia.org/wiki/Consummation
LOL :D

The Swedish word for consumption is "konsumtion", and we have no other word like it. I just picked the first recommended word FF suggested to correct my misspelling without taking a good look.
 
Last edited by a moderator:
Err, maybe I look as some reluctant, but I just confused that SW overclocking may do physical damage to CPU. Up to now I suppose that overclock can lead to SW crash (OS hangup, freeze,...) at worst.
 
peca said:
Err, maybe I look as some reluctant, but I just confused that SW overclocking may do physical damage to CPU. Up to now I suppose that overclock can lead to SW crash (OS hangup, freeze,...) at worst.
It's _very_ unlikely that something bad happens, but it's not impossible that something bad _might_ happen. That's all :)

edit:
Regarding "SW overclocking", it makes no difference if you set a physical jumper or switch a register via software. It's the same thing done differently.
 
Last edited by a moderator:
craigix said:
mali said:
Dead1nside said:
Okay so if we do any overclocking in software, without changing the voltages this will no matter what, even if it isn't the cause of any problems, void our warranty?
Nobody will be able to tell if you overclocked it or not, thus it's a non-issue ;)

_If_ it fails, you can either tell them you overclocked or you don't. It depends on you altruism coefficient. I got a replacement motherboard from Asus for my A600, because I didn't tell them.

So you're a dishonest little shit too?

I want your order number. You are banned from having a Pandora from us.

That's right, your order is canceled.
...

W
T
F

It's a feature of the system. That you're not offering warranty on it is outrageous, that you're banning mali is immature. Craig, if you hadn't lost my respect months ago and if I hadn't recently canceled, this would be it.

You can't treat your customers like s*** and be all surprised and shocked if they ever do the same to you, based on something perfectly reasonable at that. Grow up, you're leading a company, not a tree house.
 
Last edited by a moderator:
Can someone who actually knows what they're talking about answer a technical question for me please?

When the CPU is overclocked (completely disregarding overheating here), are the crashes caused by unsynchronised data in the processor? I have done some work with FPGAs before and the gate route is calculated so that the delays in the gates don't cause data to become unsynchronised at a chosen frequency.
Obviously a proper processor is not the same as an FPGA but I presume they're quite close.

Also I should just explain that when I say unsynchronised, I mean for example, a data bus not having all the pins correct due to some gate paths taking longer to propagate than others.
 
@Kyosys
1. I'm no customer of Craig anyway as ED is my contact person.
2. I'd like to speak for myself, and I don't see harm done. I don't take this forum's chatter serious.
 
mali said:
craigix said:
This issue needs to be a sticky thread, or it's just going to keep happening.
So, overclocking doesn't void the warranty and we can let this thread die?
mali said:
craigix said:
No, we can sticky it because it has been asked many times before.

An ARM CPU does not generate much heat. Overclocking it will probably just crash it if it isn't happy.

Obviously someone opening up their Pandora, hacking a voltage increase (GP32 Style) would void it. But software increases won't cause anything more than a crash.

I read that "no" as "it does NOT void warranty".
 
Last edited by a moderator:
mali said:
@Kyosys
1. I'm no customer of Craig anyway as ED is my contact person.
2. I'd like to speak for myself, and I don't see harm done. I don't take this forum's chatter serious.
Sorry mali, I didn't mean to white knight you. It's not about you, it's about Craig.
 
Last edited by a moderator:
chris_r said:
Can someone who actually knows what they're talking about answer a technical question for me please?

When the CPU is overclocked (completely disregarding overheating here), are the crashes caused by unsynchronised data in the processor? I have done some work with FPGAs before and the gate route is calculated so that the delays in the gates don't cause data to become unsynchronised at a chosen frequency.
Obviously a proper processor is not the same as an FPGA but I presume they're quite close.

Also I should just explain that when I say unsynchronised, I mean for example, a data bus not having all the pins correct due to some gate paths taking longer to propagate than others.
In short: yeah, sort of atleast. The transistors dont switch and/or the signals dont propagate fast enough (before the clock signal's next edge) => crash. I think that the basic idea doesnt change much between FPGA and an ASIC ("real processor"). It's the same thing that limit's speed even when playing with 74xx IC's...
 
Last edited by a moderator:
Kyosys said:
Sorry mali, I didn't mean to white knight you. It's not about you, it's about Craig.
Ich weiß, lass es uns einfach vergessen. Unter'm Strich bringt es sowieso nichts und zieht nur die Stimmung runter. Ich verstehe, dass Du sauer bist, aber das Projekt bleibt genial und unterstützenswert :)
 
Last edited by a moderator:
was craigix rant in jest or not?
something other than this thread (and mali) has pissed him off it that rant was serious. im hoping its not the factory...
 
Status
Not open for further replies.
Back
Top